Weather MCP Agent

A modular, production-grade AI agent that provides real-time weather forecasts. This project leverages the Model Context Protocol (MCP) pattern, Google Gemini 2.0 Flash-Lite for intelligence, and Streamlit for a responsive user interface.

Project Overview

This application serves as an intelligent agent capable of executing function calls to retrieve live meteorological data. It is built with a focus on SOLID principles, modularity, and clean architecture.

Related MCP server: mcp-foundry

Architecture

• Agent Orchestration: Uses the Google GenAI SDK to manage conversation state and tool execution.

• Weather Service: A decoupled service layer for external API communication (Open-Meteo).

• Interface: A clean Streamlit chat UI for interactive weather queries.

Project Structure

weather_mcp_project/
├── .env                  # Environment variables (API Key)
├── .gitignore            # Git exclusion rules
├── app.py                # Streamlit chat interface
├── requirements.txt      # Project dependencies
└── src/                  # Core application logic
    ├── agent_client.py   # AI Agent orchestration
    ├── config.py         # Pydantic-based configuration
    ├── mcp_server.py     # MCP tool definitions
    └── weather_service.py# External data integration